Gee, I thought you only crashed when using VASI/PAPI if you had CVD.
That's what the CASA PMO would have people believe...
There's only ever been one reported accident where a pilot's CVD was implicated which involved the crash of a FedEx B727 in Tallahassee in 2002 after flying a PAPI approach. The NTSB's evidence on the subject of colour vision came from a single source whose enthusiasm for promoting more restrictive CVD standards is legendary. Never mind the fact that there were two other colour normal pilots on the flight deck who failed to recognise the dangerously low indications that the PAPI was supposedly meant to be providing.
Dr Arthur Pape has researched this particular accident extensively and has written an article which highlights some major concerns with the 'fail unsafe' nature of the PAPI which proposes an alternative explanation of the reason of the crash.
The Puzzle of the Crash of FedEx Flight 1478: Implications for Colour Vision Standards in Aviation
There were also some interesting scientific studies published years earlier which predicted that this type of accident could occur under certain atmospheric and environmental conditions (the same conditions which were present at the time of the Tallahassee crash):
FAA Technical Center - Evaluation of Precision Approach Path Indicator - see 'Condensation tests' (pages 19-21 of 88)
Upon energization, diffusion and mixing of the projected colors created a broad "pink" signal which could not be easily interpreted. Correct color signals were restored within 15 minutes after turning the PAPI unit on as the heat from the lamps dissipated the accumulated moisture."
The pink signal, as viewed by a pilot, could have been interpreted to be white, giving a false 'fly down' signal. A dangerous condition.
FAA CertAlert 02-08: PAPI Operation
Because external PAPI lenses, used to improve light signals, are exposed to ambient weather conditions, the possibility of dew and/or frost forming on the outside of the glass becomes a concern when the units are not operated continously. In particular, a PAPI unit operating in the "off" mode for an extended period of time during the evening hours could accumulate a level of environmental contaminants (eg. dew or frost), which may not dissipate sufficiently to ensure correct light signals after pilot activation.
(Interestingly, this CertAlert was issued some 5 months AFTER the FedEx crash)
Australian Department of Defence - Hazards of Colour Coding in Visual Approach Slope Indicator Systems
Conclusions
Colour coding as a primary cue is almost universally condemned in the ergonomics literature, and an examination in this report of the numerous factors which act to degrade the reliability of colour-coded primary signals from VASIS supports the view that neither Red-White VASIS nor PAPI should be used in air transport operations. Apart from reasonably common circumstances which render colour-coded signals from these aids as unreliable. eg. atmospheric conditions and windshield and projection optics scattering, there is a strong prima facie case that the signals, through combinations of physical and physiolgical circumstances which are not rare, can become sufficiently misleading to be hazardous. T-VASIS has a shape/pattern coding as its primary cue and its colour coded extreme 'fly-up' warning signal is secondary and usefully redundant and the system is therefore failsafe in all conditions, including those in which Red-White VASIS and PAPI are hazardous. Therefore it is strongly recommended that VASIS such as Red-White VASIS and PAPI which use colour differences as the primary signal code should not be used for routine operations by military aircraft in Australia. It is suggested that Transport Australia should consider extension of this recommendation to civil aircraft routine operations in Australia and that representations should be made for ICAO to reject the use of Red-White VASIS and PAPI in international air transport operations.